Remove the PodRestarter controller and tidb.pingcap.com/pod-defer-deleting annotation (#3296)

[ti-srebot]

cherry-pick #3296 to release-1.1

---

What problem does this PR solve?

This PodRestarter is used to restart a pod by annotating it with tidb.pingcap.com/pod-defer-deleting. However, there is no user to use this feature because we recommandate users to restart pods gracefully via instrucations https://docs.pingcap.com/tidb-in-kubernetes/dev/restart-a-tidb-cluster#performing-a-graceful-rolling-restart-to-all-pods-in-a-component. The PodRestarter control will retrieve all pods belongs to the current TiDBCluster instance when every sync iteration and delete some of it, I think there is unnecessary cost.

What is changed and how does it work?

Remove the PodRestarter

Check List

Tests

• Unit test
• E2E test

Code changes

• Has Go code change

Does this PR introduce a user-facing change?:

Remove the PodRestarter controller and `tidb.pingcap.com/pod-defer-deleting` annotation

[codecov-commenter]

Codecov Report

❗ No coverage uploaded for pull request base (release-1.1@c7fd042). Click here to learn what that means.
The diff coverage is n/a.

@@              Coverage Diff               @@
##             release-1.1    #3301   +/-   ##
==============================================
  Coverage               ?   44.43%           
==============================================
  Files                  ?      148           
  Lines                  ?    14559           
  Branches               ?        0           
==============================================
  Hits                   ?     6469           
  Misses                 ?     7509           
  Partials               ?      581           

Flag	Coverage Δ
#unittest	44.43% <0.00%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.